Interpreting spacetimes of any dimension using geodesic deviation

Interpreting spacetimes of any dimension using geodesic deviation

(n:

o-

<n:

a-

a:

Jin PodolskjEI and Robert SvarcQ
Institute of Theoretical Physics, Faculty of Mathematics and Physics,
Charles University in Prague, V Holesovickdch 2, 180 00 Prague 8, Czech Republic

(Dated: January 30, 2012)

We present a general method which can be used for geometrical and physical interpretation of an
arbitrary spacetime in four or any higher number of dimensions. It is based on the systematic analysis
of relative motion of free test particles. We demonstrate that local effect of the gravitational field on
particles, as described by equation of geodesic deviation with respect to a natural orthonormal frame,
can always be decomposed into a canonical set of transverse, longitudinal and Newton-Coulomb-
type components, isotropic influence of a cosmological constant, and contributions arising from
specific matter content of the universe. In particular, exact gravitational waves in Einstein's theory
always exhibit themselves via purely transverse effects with D(D — 3)/2 independent polarization
states. To illustrate the utility of this approach we study the family of pp-wave spacetimes in
higher dimensions and discuss specific measurable effects on a detector located in four spacetime
dimensions. For example, the corresponding deformations caused by a generic higher-dimensional
gravitational waves observed in such physical subspace, need not be tracefree.

PACS numbers: 04.50.-h, 04.20.Jb, 04.30.-w, 04.30. Nk, 04.40.Nr, 98.80.Jk

<n:

u '.

>:
o,

OV

:

o:

X

I. INTRODUCTION

In the last decade there has been a growing inter-
est in exact spacetimes within the context of higher-
dimensional General Relativity, primarily motivated by
finding particular models for string theories, AdS/CFT
correspondence and brane-world cosmology. Such in-
vestigations thus concentrated mainly on various types
of black holes and black rings, see [U-IH for reviews
and further references. More general static or station-
ary axisymmetric | 9 | 4l5ll . multi-black hole Majumdar-
Papapetrou-type [16l423l | , and static solutions with cylin-
drical/toroidal symmetry (2^ - |2c| were also considered,
including uniform and non-uniform black strings (29rl36j
with the aim to elucidate their instability [37H39| . Other
important classes of higher-dimensional exact solutions
of Einstein's equations have also been studied recently,
for ex amp le Robinson-Trautman and Kerr-Schild space-
times [40l - |4a ] , extensions of the Bertotti-Robinson, (anti-
)Nariai and Plebanski-Hacyan universes (46|, higher-
dimensional FLRW-type |4?7H5l| and multidimensional
cosmological models I20luffi| (see also references therein),
specific solitons f24l l53l |54|. or various exact gravita-
tional waves — in particular those which belong to non-
expan ding Ku ndt family [55|, [56| , namely generalized pp-
waves I57H63I (for a study of their collisions see [64]), VSI

and CSI [65| spacetimes, or relativistic gyratons

Fundamental general questions concerning the classifi-
cation of higher-dimensional manifolds based on the al-
gebraic structure of the curvature tensor have been clar-
ified [z34zll) including generalizations of the Newman-
Penrose and the Geroch-Held-Penrose formalisms Ul

* podolsky@mbox.troj a. mff.cuni.cz
t robert.svarc@mff.cuni.cz

|80(. This paved the way for a systematic study of wide
classes of alg ebraically special spacetimes in higher di-
mensions [8ll484| . Investigation of asymptotic behaviour
of the corresponding fields and their global structure, in
particular properties of gravitational radiation, has also
been initiated [85|-|99(.

Nevertheless, in spite of the considerable effort de-
voted to this topic, there are still important aspects
concerning the nature of gravitational fields in higher-
dimensional gravity that remain open. Any sufficiently
general method which could be used to probe geometri-
cal and physical properties of a given spacetime would be
useful. In the present work we suggest and develop such
an approach which is based on investigation and classi-
fication of specific effects of gravity encoded in relative
motion of nearby test particles.

In fact, in standard four-dimensional General Relativ-
ity, this has long been used as an important tool for
studies of spacetimes. Relative motion of close free par-
ticles helps us to clarify the structure of a gravitational
field in which the test particles move. When they have
no charge and spin, this is mathematically described
by the equation of geodesic deviation (sometimes also
called the Jacobi equation) which was first derived in the
n-dimensional (pseudo- )Rie mann ian g eometry by Levi-
Civita and Synge potilO^ . see 11041 for the historical
account. Shortly af ter its application to Einstein's grav-
ity theory |l05l - lll4l ] it helped, for instance, to understand
the behaviour of test bodies influenced by gravitational
waves or the physical fate of observers falling into black
holes. Textbook descriptions of this equation, which is
linear with respect to the separati on vecto r connecting
the test particles, are given, e.g., in Let us also

mention that generalizations of the equation of geodesic
deviation to admit arbitrary relati ve velocit ies of the par-
ticles were obtained in the works 119l4l28j |. Further ex-
tensions, higher-order corrections to the geodesic devia-

2

tion equation, their particular applic ations an d references
can be foun d in the recent papers [T28l4T35l | and in the
monograph [l!8j .

In 1965 Szekeres [1141 ] presented an elegant analysis of
the behaviour of nearby test particles in a generic four-
dimensional spacetime. He demonstrated that the over-
all effect consists of specific transverse, longitudinal and
Newton-Coulomb-type components. This was achieved
by decomposing the Riemann curvature tensor into the
Weyl tensor and the terms involving the Ricci tensor (and
Ricci scalar). While the former represents the "free grav-
itational held" the latter can be explicitly expressed, em-
ploying Einstein's field equations, in terms of the cor-
responding components of the energy-momentum tensor
which describes the matter content. In order to further
analyze the Weyl tensor contribu tion, Szek eres used the
formalism of self-dual bivectors [l36l Il37l | constructed
from null frames. This enabled him to deduce the effects
of gravitational fields on nearby test particles in space-
times of various Petrov types. When these results are
re-expre ssed in a more convenient Newman-Penrose for-
malism [l38l Il39j | , explicit physical interpretation of the
corresponding complex scalars $ a are obtained. In par-
ticular, the Weyl scalar ^4 (the only non-trivial compo-
nent in type N spacetimes) represents purely transverse
effect of exact gravitational waves, the scalar ^3 (present,
e.g., in type III spacetimes) is responsible for longitudi-
nal effects, and *2 (typical for spacetimes of type D)
gives rise to Newt on-like de formations of the family of
test particles (see [l40k l42] for more details; inclusion
of a nonvanishing cosmological constant was described in

It is the purpose of the present work to extend these
results to arbitrary spacetimes in any dimension D > 4.
The paper is organized as follows. In section 2 we recall
the equation of geodesic deviation, including its invari-
ant form with respect to the interpretation orthonormal
frame adapted to an observer. In section 3 we perform
the canonical decomposition of the curvature tensor us-
ing Einstein's equations and the real Weyl tensor compo-
nents ^ a with respect to an associated null frame. We
thus derive an explicit and general form of the equation
of geodesic deviation. Section 4 analyses the character
of all canonical components of a gravitational field. Sec-
tion 5 is devoted to the discussion of uniqueness of the
interpretation frame, and derivation of explicit relations
which give the dependence of the field components on the
observer's velocity. In section 6 we describe the effect of
pure radiation, perfect fluid and electromagnetic field on
test particles. Final section 7 illustrates the method on
the family of pp- waves in higher dimensions. There are
also 3 appendices: In appendix A we give relations to the
standard complex formalism of D — 4 General Relativity,
and in appendix B we summarize alternative notations
commonly used in literature on D > 4 spacetimes. Fi-
nally, in appendix C the Lorentz transformations of the
*5>A scalars are presented.

II. EQUATION OF GEODESIC DEVIATION

The main objective of the present work is to investigate
and characterize the curvature of an arbitrary spacetime
of dimension D > 4 by its local effects on freely falling
test particles (observers). The gravitational field mani-
fests itself, in Newtonian terminology, as specific "tidal
forces" which cause the nearby particles to accelerate rel-
ative to each other. This leads to a deviation of corre-
sponding geodesies whose separation thus changes with
time: in various spatial directions the particles approach
or recede from themselves, exhibiting thus the specific
character of the spacetime in the vicinity of a given event.

In standard and also higher-dimensional General Rel-
ativity, such a behaviour of free test particles (without
charge a nd spin) i s described by the geodesic deviation
equation (TooMTl^

dr 2

(1)

where R ■ ot a v are components of the Riemann curva-
ture tensor, u a are components of the velocity vector
u = u a d a of the reference (fiducial) particle moving

along a timelike geodesic j(t) = {x°(r),

..,x

D-l

d.r"
(It

u ■ u — g a p u a u^ =

the parameter r is its proper time (so that
1), and Z^ are components of the
separation vector Z = Z^d^ which connects the refer-
ence particle with another nearby test particle moving
along a timelike geodesic 7(1"). The situation is visual-
ized in figure [TJ

FIG. 1. In a curved D-dimensional spacetime, nearby test
particles moving along geodesies accelerate toward or away
from each other, as given by the equation of geodesic deviation
(fl]). Here u is the velocity vector of a reference particle, and
Z is the separation vector which represents actual relative
position of the second test particle at a given proper time r.

Equation <jTJ) explicitly expresses the relative accelera-
tion of two nearby particles by the second absolute (co-
variant) derivative of the vector field Z along 7(7"),

D 2 Z»

(2)

3

in terms of the local curvature tensor and the actual rela-
tive position of the particles, described by the separation
vector Z(t) at the time r.

To be geometrically more precise, the two geodesies
should be understood as specific representatives of a
congruence j(t, z), i.e. smooth one-parameter fam-
ily of geodesies, such that 7(7-) = j(t, z = 0) and
7(r) = 7(r, z = const.). The proper time r and the pa-
rameter z which labels the geodesies can be chosen as
coordinates on the submanifold spanned by the congru-
ence. Thus u — d T and Z = d z , and the deviation vec-
tor field Z is Lie-transported along the geodesies gener-
ated by it. Consider now the positions of two test par-
ticles at a given time, for example V (located at z = 0)
and Q (for which 2 = 1, say) at r = 0, as shown in fig-
ure [TJ Their coordinates are related by the exponen-
tial map Xq — exp(zZ) xf^, generated by Z at V ', where
we set z = 1 to locate Q. If the higher-order terms are

negligible this expression reduces to Xq

(Zx^h

demonstrating that the separation vector Z describes the
relative position of the two test particles, and Z{j) gives
its evolution that is obtained by solving the equation ([TJ .
Such linear approximation improves when the second test
particle moves very close to the reference one, i.e. along
the geodesic z = const. -C 1, in which case the separation
is described by the vector field zZ(t).

It should also be recalled that the equation of geodesic
deviation (Q} is linear with respect to the components
of the separation vector, neglecting higher-order terms
in the Taylor expansion of exact expression for relative
acceleration of free test particles. It can thus be used
when the relative velocities of the particles are negligi-
ble, i.e. their geodesies are almost parallel. Generaliza-
tions of equation ([JJ) to admit arbitrary rel ative velocitie s
wer e obtained and applied in the works 1 19l4l24L ll 26l —
Il28| . Further extensions, higher-order corrections to the
geodesic deviation equ ation and their specific applica-
tions can be found in fl30MT35l (for reviews and other
references see jll8lll28lll3llll35| |). Our aim, however, in
this paper is to investigate local relative motion of nearby
free test particles which are initially at rest with respect
to each other. For such an analysis, the classical geodesic
deviation equation ([1} will be fully sufficient.

Now, in order to obtain invariant results independent
of the choice of coordinat es, it is natural to adopt the
Pirani approach [105L Il06| based on the use of compo-
nents of the above quantities with respect to a suitable
orthonormal frame {e a }. At any point of the reference
geodesic this defines an observer's framework in which
physical measurements are made and interpreted. In par-
ticular, the separation vector is expressed as Z = Z a e a .
The timelike vector of the frame is identified with the
velocity vector of the observer, e( ) = u, and em, where
i = 1,2,. ..,£) — 1, are perpendicular spacelike unit vec-
tors which form its local Cartesian basis in the hypersur-
face orthogonal to u (see also figure^,

Due to the fact that u is parallelly transported, for the
zeroth frame-component Z^ = e^ -* • Z = —u ■ Z we im-
mediately obtain

^2^(0) D 2 Z M

~Uu. ; — = -R mfiv u^u a u Z v = , (4)

dr 2

dr 2

using the skew-symmetry of the Riemann tensor. There-
fore, Z(°'(t) must be at most a linear function of the
proper time. By a natural choice of initial conditions,
consistent with the above construction of the geodesic
congruence j(t, z), we set Z^ = 0. The temporal com-
ponent of Z thus vanishes and the test particles always
stay in the same spacelike hypersurfaces synchronized
by t.

Physical information about relative motion of the test
particles is thus completely contained in the spatial frame
components Z^ 1 ' (r) = eW ■ Z of the separation vector Z.
These determine the actual relative spatial position of the
two nearby particles. By projecting the geodesic devia-
tion equation (JTJ) onto = en\ we obtain

Z® = R

(0

(0)(0)(j)

(5)

where i, j = 1, 2, . . . , D — 1, and we denote the physical
relative acceleration as

D 2 Z _ (i) D 2 Z^
"dT 2 " ~ 6 " ~d^~

(6)

The frame components of the Riemann tensor are
i?(i)(o)(o)(j) = R^apv e (i) u °' ul3e '(j)- us note that Pirani
[l05lll0d | labeled, in D — 4, the frame components of the
"tidal stress tensor" that occurs in equation ([SJ) (with
an opposite sign) as K a b = i? a o&o = R a cbd u<: u d - They
are equivalent to the electric p art o f the Riemann tensor

Sab = RaObO = RacbdU C U d , See [HI].

Following Pirani it is also usually assumed that the or-
thonormal frame {e a } is parallelly propagated along the
reference geodesic. However, in our work we do not make
such an assumption. In fact, as a key idea of the proposed
interpretation method, we align the orthonormal frame
with the algebraic structure of a given spacetime instead
(see also section [Vj) . This makes the investigation of its
physical properties much easier.

III. CANONICAL DECOMPOSITION OF THE
CURVATURE TENSOR

Next step is to express the frame components of the
Riemann tensor -R(i)(o)(o)(j)- Using the standard decom-
position of the curvature tensor into the traceless Weyl
tensor C a bcd and specific combinations of the Ricci tensor
R a h and Ricci scalar R ,

Rabcd = Cabcd + jj 2 ( i9a [ c ^ b ~ 5fc [ c '

3q/3 £a e b = Vab = diag(-l, !,...,!)

(3)

(D-1)(D -2)

Rg a [ c 9d]i

(7)

4

we immediately obtain

%(o)(o)(j) = C(i)(o)(o)(j) + Jj—^( R (m ~ Si i R (0)(0))

RSu

(D-1)(D -2)

(8)

Before substituting this into the geodesic deviation equa-
tion ([5]) we also employ the Einstein field equations, gen-
eralized to any dimension D > 4 ,

Rab — hRgab + ^ 9ab — 8lT T ab

(9)

where A is a cosmological constant and T ab is the energy-
momentum tensor of the matter field. Using ([9]) and its
trace R = {8ttT - DA), we rewrite © as

Using the notation of [93| , the components of the Weyl
tensor in such a null frame are determined by the follow-
ing scalars (grouped by their boost weight):

*o« = C abcd k a m\ k c m d ,

= C abcd k a m\ m c } m d , f 1T< = C abcd k a l b k c m d ,
* 2 «« = Cabcd ml m) m c k mf , * 25 = C abcd k a l» l c k d ,

* 2 « = C a6cd k a l b m^m d

2T»J

* 3 «* = C a6ca Z Q m^ , * 3T , = C abcd l a k h l c m d ,

*4« = C abcd l a m b i l c m d

(14)

where i,j,k,l = 2,...,D—l. All other frame compo-
nents can be obtained using the symmetries of the Weyl
tensor. The scalars in the left column are independent,
up to the obvious constraints

i? (i)(0)(0)(j)

2A(5ii

(D-l)(D-2)
8tt

+ CVi

(i)(0)(0)(j)

(10)

D - 2

T (i)G) _ 5 ij ( T (0)(0) + J^—j)

The equation of geodesic deviation ([5]) thus takes the
following invariant form

2A

(D-l)(D-2)
8tt

Z® + C (i

(0(0)(0)(j)

Z (j)

T (i)(j) Z(l) - ( T (0)(0)

L> - 1

(11)

The first term represents the isotropic influence of the
cosmological constant A on free test particles, the second
term describes the effect of a "free" gravitational field
encoded in the Weyl tensor, while the second line in (jTTJ)
gives a direct effect of specific matter present in a given
spacetime.

The terms proportional to the coefficients C(;)(o)(o)(j)
can further be conveniently expressed using the
Newman-Penrose-type scalars, which are the compo-
nents of the Weyl tensor with respect to an associated
(real) null frame {k,l,m,i}. This frame is introduced by
the relations

k

m,i = eu\ for i = 2, . . . , D — 1

Hi))-

I

(12)

where u = is the velocity vector of the observer.
Thus k and I are future oriented null vectors, and
are D — 2 spatial Cartesian vectors orthogonal to them,
satisfying

k ■ I = — 1 , mi ■ rrij = 6ij ,

k ■ k = = I ■ I , k ■ rrii =0 = 1- mi . (13)

= o,

= 0,

= 0,

= 0,

\I> 2 yk!

= 0,

= v I / 2 >[j'=i] =

o,

-o,

= 0,

*4«]

= 0,

#4**

= 0,

(15)

while those in the right column of (I14[) are not indepen-
dent because they can be expressed as the contractions
(hence the symbol "T" which indicates "tracing" )

^ 1T i = *

^>2S= v E , 2T fcfc = 5*2 fci '

(16)

*2T«= §(*2*«* +*2«)

where * 2 t(«) = |*2*k/ , * 2 tW1 = |*2«

In the case D = 4, these Weyl tensor components i n the
null tetrad reduce to the standard Newman-Penrose jl38l .
139] complex scalars ^a- Explicit expressions are given
in appendix lAl

Using relations e (0 ) = -^(k + I) , e (1) = -^=(k - I)
and the definition (TT4"|) . a straightforward calculation
then leads to the following expressions for the compo-
nents C(i)(o)(o)(j) of the Weyl tensor which appear in
equation (fllj):

C(l)(0)(0)(l) = *2S )

C(l)(0)(0)(j) = y= (*lTi - *3T^) ,

C(i)( )(0)(l) = -/= (*1T< - *3T0 ,

(17)

C

«(0)(0)(j)

"2 (*o« '

2T( i J) 1

where i,j = 2,...,D — 1 label the spatial directions or-
thogonal to the privileged spatial direction of em.

5

Putting this into (fTTj) we obtain the final invariant and fully general form of the equation of geodesic deviation:
2A

(D-1)(D-2) Z{1)+ * 2S Zil) + 7=2 ( * 1W " * 33 " } ZW

8tt
D-2
2A

T ( x) (1 ) z« + t (1)(j) z<*> - (t (0)(0) + r)

(D-1)(D -2) 2T ' V2 2 V 4 7

8tt
D-2

T ( . i)(1) z (1) + T (i)(i) Z«> - (r (0)( o) +

L> - 1

T Z

(i)

(18)

(19)

This completely describes relative motion of nearby free test particles in any spacetime of an arbitrary dimension D.
In the next section we will discuss the specific effects given by particular scalars which represent the contributions
from various components of the gravitational and matter fields.

I

Finally, we remark that our notation which uses ^a
in any dimension is simply related to the notations em-
ployed, e.g., in [zl [zl], in [zl, HH and recently in [79j .
The identifications for the components present in the in-
variant form of the equation of geodesic deviation are
summarized in table HI More details are given in ap-
pendix [B] in particular see expressions (|B8I) , (|B11|) and
l|B18jl .

TABLE I. Different equivalent notations used in the literature
for the Weyl scalars which occur in the equations of geodesic
deviation (fill), ED).

refs. [72, 73|

refs. [76, 81]

ref. [JJ

— Coioi

_$

_$

*2T«

—Conj

*1TJ

— Coioj

Cioij

*;

*0«

C'oiOj

#4«

2*«

IV. EFFECT OF CANONICAL COMPONENTS
OF A GRAVITATIONAL FIELD ON TEST
PARTICLES

Let us consider a set of freely falling test particles,
initially at rest relative to each other, which form, e.g.,
a small (hyper-) sphere. In any curved spacetime, such a
configuration undergoes tidal deformations which can be
deduced from the accelerations measured by the fiducial
observer attached to the reference test particle in the
center. The resulting relative motion represents the effect
of a given gravitational field, whose specific structure is
explicitly characterized by the system (fT5)l , (|19l) .

First concentrating on the vacuum case, i.e. T a b = 0,
the system of equations describing purely gravitational

interaction simplifies considerably to
2A

z« =

Z^+^oaZ^ 1

(D-l)(D-2)

1

7=2 ( * 1TI

3T' ;

z m

(20)
(21)

The overall effect of the gravitational field on test parti-
cles is thus naturally decomposed into clearly identified
components proportional to the cosmological constant A
and the Weyl scalars ^a ■ Of course, for algebraically
special spacetimes some (or many) of these coefficients
vanish completely, and even in algebraically general cases
specific numerical values of the scalars ^a can distin-
guish the dominant terms from those that are negligible.
Let us now briefly describe the character of each term
separately, including its physical interpretation.

• A : isotropic influence of the cosmological back-
ground

The presence of the cosmological constant A is en-
coded in the term

( ZW \ _ 2A ( 1 Q \( ZW '

1 z(0

(D-1)(D -2) 1 6,

Z U)

(22)

which

can be

2A

written in

unified way
£?M for all spatial compo-
1. In parallelly propagated

(D-l)(D-2

nents i = 1, 2, . . . , D
frames, this yields the following explicit solutions

A =
A >

At t + Bi ,
Ai cosh

Z (0

+Bi sinh
A < : Z (i) = Ai cos
+B[ sin

2A

(£>-l)(2?-2)

2A ~

(£>-!) (D-2)

g|A]

C-D-1XZJ-2)

gJAj

(D-l)(D-2)

6

where A u Bi are constants of integration. These
are characteristic relative motions of test parti-
cles in spacetimes of constant curvature, namely
Minkowski space, de Sitter space and ant i-de Sitte r
space, respectively, as derived by Synge [l02l Il03l ].

• ^4,3 : transverse gravitational wave propagating in
the direction +em

This part of a gravitational field influences the test
particles as

(fw) = ~\{l (z«)' (23)

Obviously, this is a purely transverse effect because
there is no acceleration in the privileged spatial di-
rection em. The set of scalars forms a sym-
metric (vE^ij = \E , 4ji) and traceless ( 1 i 4 k k = 0) ma-
trix of dimension (D — 2) x (D — 2), cf. the last
line in (TT5|) . so that it has \D(D — 3) independent
components corresponding to polarization modes
(see also [ll, [lH U^)- In direct a nalogy w ith a
linearized Einstein gravity in four jll5l Ill6l ] and
higher dimensions [86, 93], ^4^ represents the grav-
itational wave which propagates along the null di-
rection fc, i.e. in the spatial direction +em (in view
of relations (TT21 there is fern = k ■ em > while
km = k ■ e(j) = for i = 2, . . . ,D — 1). Spacetimes
of algebraic type N (for which only the components
^ 4 i3 = Cuij are nonvanishing [7^[73|]) can thus be
interpreted as exact gravitational waves in any di-
mension D > 4.

• ^ 3T i : longitudinal component of a gravitational
field with respect to +em

Such terms cause longitudinal deformations of a set
of test particles given by

[ z« J V2\ * 3T . J { z& J ■ 1 j

These (D — 2) scalars <I/ 3T i , which combine motion
in the privileged spatial direction em with mo-
tion in the transverse directions e^, are also ob-
tained using ^ 3T i = ^> 3 k k i, where 4/ 3 ijk = — ^ 3 ikj
and tyyjk + ^ 3 jki + *S> 3 kij = 0. Longitudinal effects
of this type occur in spacetimes of type III and in
algebraically more general cases.

• ^2S, ^2T(«) : Newton-Coulomb components of a
gravitational field

The terms

( f « ) = ( *o -* 2 ° rW) ) ( zw ) (25)

give rise to deformations which generalize the clas-
sical Newton-Coulomb-type tidal effects in D = 4,

namely those in the vicinity of a spherically sym-
metric static source. Recall that vE^s = ^t**
(see (fl~6|) and (p~5)) for further relations), so that
the (D — 1) X (D — l)-dimensional matrix in (|2"5)l
is symmetric and traceless. These terms are typ-
ically present in type D spacetimes, for which the
notation = — $ and — ^2T(^> = is com-
monly used [Zi, [7HH ill, see (IbTT]) . As shown
in (|A6I) . the only nonvanishing coefficients of this
type in four dimensions are the diagonal elements
i* 2S = * 2T(2 2) = *2T(33> = -Re* 2 ■

• ^i T i : longitudinal component of a gravitational
field with respect to —em

The corresponding effect on test particles is

{ zw ) V2 ^ * 1T , ) { z& ) ' [Zb)

which is very similar to the acceleration caused
by the longitudinal component ty 3T i , as de-
scribed by (|24l) . In fact, it is its counter-
part: it follows from the definition (Ti"4"]) that
the scalars = ty 1 k k i (where ^S^jk — — SE^*^

and ^ink + "fpfc; + ^ihij — 0) are equivalent to
ty 3T i under the interchange k ■<-» I. Since
fcm = fc • em > while im = Z • em < 0, the
scalars ^> 1T i represent the longitudinal component
of the field associated with the spatial direction
-e ( i).

• VE'om : transverse gravitational wave propagating in
the direction —em

This component of a gravitational field is charac-
terized by

(fw ) = ~2 (0 *1 ) (z« ) ' (2?)

which is fully equivalent to (|23|) under fc -H> Z. The
scalars ^gij (which form a symmetric and traceless
(D - 2) x (D - 2) matrix: * « = <f 0J! , * Qfc fc =
0) thus describe the transverse gravitational wave
propagating along the null direction Z, i.e. in the
spatial direction — em. Superposition of gravita-
tional waves which would propagate in both direc-
tions simultaneously (that is an "outgoing" wave
given by ^4^ and an "ingoing" wave given by V^o^ )
can only be present in spacetimes which are of al-
gebraically general type.

V. UNIQUENESS OF THE INTERPRETATION
FRAME AND DEPENDENCE OF THE FIELD
COMPONENTS ON THE OBSERVER

The canonical components of a gravitational field de-
scribed in previous section are represented by the real co-
efficients ^ a ■ These are projections of the Weyl tensor

7

onto particular combinations of the null frame {fc, I, rrii},
as defined in (|14l) . They are spacetime scalars and in this
sense the above physical interpretation is invariant. On
the other hand, the values of 'Fa depend on the choice
of the basis vectors of the frame. In this section we will
argue that such a dependence corresponds to simple local
Lorentz transformations related to the choice of specific
observer in a given event, and that the natural interpre-
tation null frame is essentially unique.

Let us consider an observer attached to the reference
(fiducial) test particle moving through some event in the
spacetime, such as the point V in figure [TJ whose ve-
locity vector is u. This timelike vector (normalized as
u ■ u = — 1) defines an orthogonal spatial hypersurface of
dimension D — 1 spanned by the Cartesian vectors en\,
where i = l,2,...,D — 1. Assuming the spacetime is of
an algebraic type I or more special, it is most natural to
associate the corresponding Weyl aligned null direction
(WAND) with the null vector fc of the interpretation ref-
erence frame, see figure O

FIG. 2. Natural choice of the interpretation null frame and
the related orthonormal frame (|12p . (|13|) . Up to spatial ro-
tations of rrii = efj), they are uniquely given by the velocity
vector u of the observer and the WAND k at any event V of
the spacetime.

The privileged unit vector em, defining the longitudinal
spatial direction, is then uniquely obtained by projecting
k onto the spatial subspace orthogonal to u. This also
fixes the normalization of k (to satisfy the first relation
in (|12l) we require fe • u = — -^=). The complementary
null vector I of the frame is then also uniquely given via
the relation I — \J~2u — fe. It only remains to choose the
transverse spatial vectors ep), . .., em-i)> he. m-i = &u)-
As shown in figure [21 these must lie in the (D — 2)-
dimensional subspace orthogonal both to u and em, so
that fc • rrii =0 = 1- rrii as required by ([TBI . Neglect-
ing possible inversions, the only remaining freedom are
thus standard spatial rotations represented by the rota-
tion group SO(D — 2) which acts on the space spanned

by rrii, see the explicit relation (IC4[) presented in ap-
pendix o

For any spacetime of type N (in which the WAND has
maximal alignment order) the null vector fc is unique.
In spacetimes of other algebraic types (namely IIL, Ilj,
L and D) different WANDs exist. These can alterna-
tively be used as the vector fc of the interpretation null
frame {fc, I, rrii}. Because the distinct WANDs can al-
ways be related using the null rotation with fixed I, as
given explicitly by equation (|C2j) in appendix [Cl it is
straightforward to evaluate the "new" values of the Weyl
scalars ^a using the expressions (??). Notice that the
coefficients ^4^, which are the amplitudes of transverse
gravitational waves propagating along fc, are invariant
under such a change.

Let us now consider another observer moving through
the same event V with a different velocity u. Locally,
this transition is just the Lorentz transformation from
the original reference frame {e a } to {e a } for which

(28)

where v\, . . . , vd-i are components of the spatial veloc-
ity of the new observer with respect to the original Carte-
sian basis em. This can be obtained as the combination
of a boost in the fc — I plane followed by a null rotation
with fixed fc, see equations (|C3[) and (IC1|) in appendix [Cl
if we take the specific parameters

1

B = , Li =

l-Vx

where i = 2, . . . , D — 1:

1 v^JJ-i 2

i-E i= i «i

(29)

fc

1

fe,

D-l

Vi rrii

I -Vl

(30)

rrii = mi

V2-

1-vi

Indeed, u=A^(k + l) gives exactly the relation (|28l) .
The corresponding change of the Weyl scalars can
thus be obtained by combining (|C7[) with (|C5I) . which
yields

8

1 ~

= - 2 v / 21< 0l[3 A fc] ,

u

* 2 y W = fr 2yw _ 2^ (X^fcjy - + 4 (^[ftX^X; + * oj[! A fc] A 4 ) ,

*2S = *2S - 2\/2* 1Tl X J - 2>I'., A '.V ,
# 2 « = + v / 2*ife«X fe - 2v / 2* 1T [«X 7 -] - 4* oM i^O]X* ,
* 2rw = <& 2Ty + V2*p fcJ X fe - V2V 1T iXj - 2^ , k X k X 3 + * ^l^| 2 , (31)

+4$ 1Tlj X k] Xi - 2 (* uu X k + VujnXi - ^kuX^X 1 + ^^ 3k \X\ 2
+iV2^ Qlu X k] X l X l - 2V2V oHl X k] \X\ 2 ,

B § 3T « = * 3T , + V2 ^2'oX 3 - V2 (*S>2T k iX k + ^2SXi)

+2 (2* 1T3 Xi - ^ lk ^X k ) X j - V 1T <\X\ 2 + 2\ 2v[',. . A •'.V / '.V, - v / 2# «X»|.X'| a .
B 2 * 4 y = + 2V2 ($ 3T( ,X 3) - * 3(Ij)fc A fc )

+2* 2lfc3i A^A* - 4* 2Tfc(l A i) A fe + 2* 2T(I3 , |A| 2 - 2f 2 , s A% : A i - At> 2 H*X ]) X k
- 2V2{2m im ,X j) X k X l + V imk X k \X\ 2 + # 1T(i J^)|X| 2 - 2# 1Tfe X fc X;Xj)
+4* oW A fc A w A i Aj -4^ Qk{ ,X f) X k \X\ 2 + 4< 0I3 |A| 4 ,

where we denoted

X, = BU = — ^— . (32)
1 - Ui

In particular, for spacetimes of algebraic type N, which
admit a WAND of the maximal alignment order, the only
nonvanishing component of the gravitational field is
representing the transverse gravitational wave propagat-
ing in the spatial direction e^y It immediately follows
from (|3"Tj) and (|29[) that the transition to any other ob-
server results just in a simple rescaling of the gravita-
tional wave amplitudes

i-E i= i «?

4'.?

(33)

If the new observer moves only in the spatial direc-
tion in which the wave propagates, V\ > and Wj =
for i = 2,...,D- 1. Then * 4 « = (1 - «i)/(l + «i) * 4 y
which is smaller than If the observer's velocity ap-

proaches the speed of light, V\ — > 1, the amplitudes of the
gravitational wave disappear, ^ ^i, — > 0. Contrary, when
the observer moves against the wave its amplitudes grow,
and for V\ — > — 1 they diverge.

by the invariant form of the equation of geodesic devia-
tion (JTHJ) , (|19[) . Setting the cosmological constant A and
all components of the Weyl tensor to zero, it reduces to

87T

D

8tt
D — 2

T {im ZW+T {im ZW
2

r W(D

T (o)(o) I" jj—j
2

T,

(0)(0)

D - 1

T) Z {i)

(34)

It will be illustrative to investigate some important types
of matter usually contained in the families of exact so-
lutions of Einstein's equations, namely pure radiation,
perfect fluids and electromagnetic fields.

VI. THE EFFECT OF MATTER ON TEST
PARTICLES

Let us now consider the direct effect of specific forms of
matter on relative motion of test particles, as described

9

• pure radiation

The energy-momentum tensor of a pure radiation
field (or "null dust") aligned along the null direc-
tion k is

Tab — pk a kb ,

(35)

where p is a function representing the radiation
density. Its trace vanishes, T = 0, and using (fl2|)
we derive that the only nonvanishing components
of T a b in the equation of geodesic deviation are

T,

(0)(0)

(i)(i)

considerably to

Air p
D-2

Equations (1M)) thus reduce

(36)

In an arbitrary dimension D there is thus no ac-
celeration in the longitudinal spatial direction eny
The effects in the transverse subspace are isotropic
and (since p > 0) they cause the radial contraction
which may eventually lead to an exact focusing.

• perfect fluid

For a perfect fluid of energy density p and pressure
p (which is assumed to be isotropic) the energy-
momentum tensor is

Tab = (p + p)u a u b +p g a b

(37)

Provided the fluid is comoving, its velocity u coin-
cides with the observer's velocity which is the vec-
tor e( ) of the orthonormal frame. The trace is
T = (D — l)p — p, and the relevant nonvanishing
frame components are T/ \/m = p, Tmm =P and
=pSij. The equation of geodesic deviation
thus takes the form

-8tt

{D-3)p+(D-l)p ( 1

(D- !)(£>- 2)

S.

The resulting motion is isotropic, the same in the
longitudinal and all transverse spatial directions.
For positive p and p, the fluid matter causes a con-
traction, such as in the case of dust (p — 0), inco-
herent radiation (p = -^5y p) , or stiff fluid (p = p).
However, for matter with a negative pressure, the
set of test particles may expand. In particular, if
the matter is described by the equation of state
p = —p = const., it mimics the cosmological con-
stant A = 8irp since (J3HJ) is then completely equiv-
alent to (1221).

• electromagnetic field

The energy-momentum tensor of an electromag-
netic field is given by

-If

FacF b C

1

■ gab F c d F'

cd

(39)

so that its trace is T = ^(4 - D) F ab F ab . The
frame components of T a b which occur in expressions
(1341) are

T (o)(o) = ^ (F{q) c F (0) c + -F ab F ab ^j ,

J (l)(l) - 4^(1)^(1) -^abt ),

T (1)W - ^ F (l)c F « C >

(40)

f

T M(J) - 4^ ( F (i)c F U) C - 1 5 ij F *b F ab

In this case the equation of geodesic deviation takes
the following more complicated form:

) \TiTij)\

(41)

where

t=^(f (1)c f {1) c ~f (0)c f {0) c )

3

(D-l)(Z?-2) °
o

F{i) c F(i

F nh F

(lb

D-2
2

D-2

(42)

3

$ij F a b F a .

(D-l)(D-2)

We observe that the clear distinction between the
longitudinal and transverse spatial directions is not
present, except at very special situations. Some im-
portant particular subcases can be easily identified
and analyzed, for example a null electromagnetic
field for which the invariant vanishes, F a b F ab = 0,
or purely electric aligned field in the vicinity of
static black holes.

VII. AN EXPLICIT EXAMPLE: PP- WAVES IN
HIGHER DIMENSIONS

We conclude this paper by demonstrating the useful-
ness of the above interpretation method on an impor-
tant family of exact spacetimes, namely the pp- waves.
These are defined geometrically as admitting a covari-
antly constant null vector field k. Such CCNV space-
times thus form a special subclass of the Kundt space-
times because the geodesic congruence generated by k is
twist-free, shear-free and non-expanding.

In [55[ we investigated general Kundt spacetimes in
higher dimensions, admitting a cosmological constant A
and a Maxwell field aligned with k (which is necessarily
a multiple WAND). In natural coordinates the metric of

10

all such pp-waves can be written in the Brinkmann form

of geodesic deviation (|T8| and (fT9|) :

ds 2 = gij dx l dx j + 2 a dx l du - 2 du dr + c du 2 , (43)

where k oc d r and g^, e;, c are functions of the transverse
spatial coordinates x k and the null coordinate u. The
explicit Einstein-Maxwell equations can be found in [55[ ,
namely equations (115)— (118).

For the metric (1431) the interpretation null frame
adapted to a general observer which has the velocity
u = r d r + u d u + x 2 d x 2 + . . . + x D ~ 1 d x n-i is

8 r ,

V2ii

I = (V2r - -^-) d r + V2iid u
V y/2u>

+V2x 2 d x 2 + ... + V2x D - 1 d x n-i , (44)

m t = -r{eku + g k x?)m k d r

-rn l d x 2

where gki rn\ rrA = Sfj , and nontrivial components of the
Weyl tensor are

Crum- (jD _ 1)(jD _ 2 ) Si? '

Criuj- D _ 2 B Ri j {D _ 1){D _ 2)

( r

i

R,

1

Cijhl = S Rijkl

Cuijk — Ruijk

Ciuju — Riuju

D-2 (D-l)(D-2)
2

s Rg i3

s Re l

D-2

{gi[k s Ri} 3 - gj[k s Ri]i)

(D-1)(D -2)

s Rgt[k9i] 3 , (45)

D-2

( e [j S Rk]i ~ 9i\j Rk]u)

(D-1)(D -2)

S R e [j 9k]i >

1

D-2

(c Rij 2e^i?^ u -|- Qij Ruu)

l

(D- !)(£>- 2)

s R(cg l3 - ei e 3 )

Using definition (fT4[) we evaluate the Weyl tensor (|45l) in
the interpretation null frame (|4"4"]) . Lengthy calculation
(with some "miraculous" cancelations) gives the follow-
ing nonvanishing Weyl scalars which enter the equations

*2S
*4«

1

(£>-l)(£>-2)

so k I

Rki m, m

'R,

D-2
V2

s RSi

D-2

l "" J (D — 1)(D — 2) "" v '

( s R km x m + R ku u)m k , (46)
1

mlu fZ n 9kl Rmu ) % U

Rkulu ~ -jj ^ 9kl Ruu )

This is a general result valid for any pp-wave spacetime
because no particular field equations have not yet been
imposed.

Notice that ^ 2 t^ — ^2T(n) ■ Moreover, in accordance
with the relations Ql]) and (jTSj , * 2 s = *2T*= fc and
1 $>< l k k = so that any pp-wave is traceless.

The relative tidal motion of nearby test particles in
general pp-waves will thus be caused by the combina-
tion of the transverse gravitational wave (|23p propagating
along k with amplitude \P4sj , the longitudinal component
(PMl) of the gravitational field with amplitude \E , 3T i , and
the Newton-Coulomb contribution (|25[) determined by
the scalars vE^s and 1 $ , 2 T i i'-

V2

(47)

There is also the isotropic background influence ([22]) if
the cosmological constant A is present, or the interaction
(|4"T1) with the electromagnetic field.

The scalars (|46|) which enter (|47|) combine kinematics
(namely the velocity components x m , fi of the observer)
with the specific curvature of spacetime encoded in the
only nonvanishing components of the Riemann and Ricci
tensors, namely

Riiki = S R

R

Hjkl
uijk

ijkl

\ ( e fc,

ik ~l~ 9ij,uk 9ik,uj^)

<~ L ij\ 2 9km.u \ V[m,k\) 1 ik \ i^jra.u ~r ^[m,^] } •>

- s r^.(e fe , u -ic fe ),

11

and

Rij Rij '

W k (\9i3,u + e {jA )]^ k

+g 3k g lm (gim.k e[i,j\ - \ghm,i9ji,u) - (lnV?),™ >

Run = -W ic dh - W ic ,o){^V9),i (49)

l) .1

(lnV5),

-\9 ij 9 U 9ik,u9jl

g l3 g kl e[i yk ] ej.i

where s Rijki and s Rj denote, respectively, the Riemann
and Ricci tensors corresponding to the spatial metric gij
only. The Ricci scalar S R (equal to R) of this transverse
(D — 2)-dimensional Riemannian space enters, in fact,
only the Newton-Coulomb scalars "I^s and 5 , 2 T i J ■ Inter-
estingly, these are also independent of the velocity of the
observer.

There is a big simplification if we restrict ourselves
to vacuum pp- waves. As shown in [55[, the absence of
an aligned electromagnetic field requires that the cosmo-
logical constant A also vanishes, so that the transverse
Riemannian space must be Ricci flat, s Rij = 0. In such
a case = = ^2T» ■ Moreover, since R u = = R uu ,
the Weyl scalar ^> 3T i also vanishes and the gravitational
wave amplitudes reduce to

¥ 4 « = 2[ s R kmln x m x n -
x

ZRkmlu + Rkulu U 2 ]

(50)

Taking the simplest possibility of a flat transverse
space,

9H = % . (51)

we obtain an important family of exact vacuum plane-
fronted gravitational waves (possibly representing an ex-
ternal field of gyratons [gg, 167} ) which propagate in
Minkowski space. In fact, these metrics with

S Rijkl
Ruijk

R

IUJU

o,

2 ( e k,ij
"i,uj

(52)

= |(e

- C,y) + 8

tel.

3[fc,i] e [Uj] ,

belong to the family of VSI spacetimes [63j .

If the functions can be globally removed by a suitable
coordinate transformation (in the absence of gyratonic
sources) the metric reduces to

(53)

ds 2 = Sij da; 1 da; 3 - 2 du dr + c(x k ,u) du 2

In such a case the spatial vectors of the null frame (|44j)
are simply m; = (i l /u) d r + d x i, and the frame is paral-
lelly transported. This implies that the physical relative
accelerations ([6]) are, in fact, ordinary time-derivatives of
the components of the separation vector, Z^> = -^Z^ .
Moreover, u = const, along the geodesic since there is
5T

r% = for the metric

The scalar components of the gravitational field
5"2f simplify to

=

-u 2 c ,

(54)

Using (I49[) . the only remaining Einstein's vacuum equa-
tion R uu = reads A c = 5 lJ c.ij = which explicitly
guarantees that the (D — 2) x (D — 2) symmetric matrix
of the wave amplitudes tt^y is traceless. The equations
of geodesic deviation thus reduce to

d 2 ZW
dr 2

d 2 Z^
dr 2

0,

(55)

exhibiting the transverse character of the vacuum grav-
itational pp- waves propagating along em. In general
there are \D{D — 3) independent polarization modes
corresponding to the same number of free components
of the matrix vE^ij .

In particular, if the metric function c is a quadratic
form of the transverse spatial coordinates,

D-l

where the constant coefficients Ai must satisfy

D-l

5^=0

(56)

(57)

i=2

^4*3 is a traceless diagonal matrix with eigenvalues
\l/ 4 ij = ~2AiU 2 . The amplitudes are constant, i.e., the
corresponding gravitational waves are homogeneous. If
the test particles are initially at rest [Z^'(t = 0) = 0,

0)

const.], equations of geodesic devia-

tion ([55]) for (|55|) can be explicitly integrated to

Z,

(i)

Zf c.b

7 {i)

COsh (\fA~i \u\ Tj

3(A/=A|w|r)

for
for
for

A l > ,

A t < , (58)

A> = 0.

Therefore, in the transverse spatial directions with
Ai > the test particles recede, while in those directions
with Ai < they focus. There is also a possibility that
Ai = 0, in which case there is no influence of the gravi-
tational wave in the corresponding transverse spatial di-
rections.

This results in completely new effects which are not
allowed in classical D = 4 General Relativity for which
i = 2,3 and the constraint (|57|) is simply A2 = —A3-
Therefore, either a vacuum gravitational pp-wave in 4-
dimensional spacetime is absent {A2 = —A3 = 0), or
it generates specific particle motions in both transverse
directions e( 2 ) and e( 3 ) (focusing in one of them). In
higher dimensions, however, the amplitudes are coupled

12

via the D-dimensional constraint A2 = —A3 — Xt^ 1 ^i-
From the point of view of a detector located on a (1+3)-
dimcnsional brane with spatial directions em, £(2)> 6(3)
this would clearly exhibit itself as a violation of standard
TT-property of gravitational waves (unless X)^!^ =
which corresponds to a very special subcase). Such an
anomalous behavior could possibly serve as a sign of
the existence of higher dimensions (see also discussion
of a similar effect within the context of linearized 5-
dimcnsional gravitational waves |91j).

It may also happen that Ak = for some k (in which
case the metric function c given by ((5*6)) is independent of
the corresponding spatial coordinate x k ) and thus there
is no effect of the vacuum gravitational pp-wave on test
particles in the transverse spatial direction bqa . Even
the special situations with A2 = or ^3=0 are allowed.

VIII. CONCLUSIONS

Let us co nclud e this work by quoting from the classic
monog raph [Hi, page 35: "In Einstein's geometric the-
ory of gravity, the equation of geodesic deviation sum-
marizes the entire effect of geometry on matter." This
is true not only in standard D — 4 General Relativity,
but also in its extension to any higher number of dimen-
sions. Indeed, we have explicitly demonstrated that the
geodesic deviation equation, expressed in a suitable refer-
ence frame adapted to the observer's geodesic and to the
specific algebraic structure of a given spacetime, can be
used as a useful tool for analysing and understanding the
specific effects of the gravitational field in an arbitrary
dimension.

In particular, we derived the general canonical de-
composition (USJ) , (JTj3J) of relative accelerations of nearby
test particles freely falling in any spacetime. The grav-
itational contributions, identified and described in sec-
tion IIV1 consist of the isotropic background influence
(f2"2")l of the cosmological constant A, transverse gravita-
tional waves (|2"3")l . (|2T|) . complementary longitudinal ef-
fects ([24]) , ([26]) . and the Newton-Coulomb component
(|2"S")) of the gravitational field. The matter contributions
were discussed in section IVI1 namely the influence of a
pure radiation field (null dust) (f551) . perfect fluid (f3"5]l.
and generic electromagnetic field (|4"Tj) .

In the final section IVIII we also exemplified these re-
sults on an important family of exact pp- waves in higher
dimensions (admitting a covariantly constant null vec-
tor field fc). Their nontrivial amplitudes are given by
expressions (|4"""T[) . The vacuum VSI subclass of such
Kundt spacetimes represents purely transverse gravita-
tional waves propagating along the WAND k (in general
associated with gyratonic sources). These exact gravita-
tional waves have amplitudes 4-4»j determined by equa-
tions (f50]). (|52l) . which form a (D - 2) X (D - 2) symmet-
ric traceless matrix. Its \D(D — 3) components charac-
terize the independent polarization modes. Explicit so-
lution of the invariant equation of geodesic deviation for

the metric function (|56|) is given in (|58| . Due to coupling
between the eigenvalues of 4- 4.3 , such higher dimensional
gravitational waves could possibly be identified obser-
vationally in (l+3)-dimensional brane as a violation of
standard TT-property.

We hope that the presented general method of inter-
preting exact spacetimes, based of the study of geodesic
deviation, will help to elucidate the physical and geomet-
rical properties of various explicit solutions of Einstein's
equations in an arbitrary dimension.

ACKNOWLEDGMENTS

The work of J. P. was supported by the
grant GACR P203/12/0118 and by the project
MSM002 1620860. R. S. was supported by the grants
GACR 205/09/H033, SVV-263301 and GAUK 259018.

Appendix A: Relation to complex notation in D = 4

In standard D = 4 General Relativity it is usual — in-
stead of the real null frame {fc, I, m.2, m.3} — to introduce
a complex null tetrad {fc, l,m,rh} and to parametrize
the Weyl tensor by the corresponding five complex com-
ponents. Thes e Ne wman-Penrose scalar quantities 4- ,4,
first defined in jl38l | , are closely related to the real quanti-
ties introduced in our text. Here we present a dictionary
relating these two notations. In D = 4 the transverse
spatial index i runs only over two values 2, 3 and we can
combine the real vectors rrii into the complex vectors

m = ^ (m 2 - im 3 ) , m = ^= (m 2 + i m 3 ) . (Al)

Any real spatial vector V spanned on , m.3 can be
parametrized by a complex number V via the relation

V = V 2 m 2 + V 3 m 3 = ^(Vm + Vrh) , (A2)

so that V — V 2 — iV 3 and \V\ 2 = (V 2 ) 2 + {V 3 ) 2 = VV.

In four dimensions there are only two real indepen-
dent components of the Weyl tensor for each boost weight,
namely

4-Q 22 = — 4-Q33 1 5*023 =

4- 17*2 = 4-4.332 = — 5^323 ,
4-22323 ~ ^23232 — — ^2 3223

— 2 ^2T 33

= 4- 2S ,

4- 2 23

= -* 2 32

= 2 ^2T 23

4- 3T 2

= 4 3 332

= -'3/3323 ,

4^22

= -* 4 33

4-423

4-Q32 ,

^1T 3 — Wl223 = — ^1 232 ,
= — \E f 22332 — 2 ^2T 22

— 2 ^2T 32 7

^3T 3 = ^3 22 3 — — y/3232 ,

4-432 . (A3)

These can be combined into five complex NP components

13

[TH [HI defined by

*o = C abcd k a m b k c m d
^i=C abcd k a l b k c m d ,
*2 = C abcd k a m b rh c l d ,

^3 = C abcd l a k b l c fh d ,

*4 = C ab cdl a rh b l c fh d ,

(A4)

According to (JTTJ) , the orthonormal components
C(i)(o)(o)(j) of the Weyl tensor are

as

*i = 75(*it 2 -i*it»)

* 2 = -|(*2 2323 +i*2 23 )
*3 = 7f(*3T 2 +i*3T 3 )

* 4 = * 4 22 + i ^ 4 23 .

(A5)

Notice the differences with respect to the notation used
in [9^: here we have re-labeled all transverse spatial
indices as + 1 to achieve that the privileged spa-

tial direction is denoted as mi = em, and the scalars
^2S, ^E , 2T i J are defined in (IT41 without the unnecessary
factor 2. (Also, there is a missing factor | in equation
(A. 7c) in [9j].) Inversely, we obtain

* 22 = -* 33 = Re* , *o 23 = *o 32 = -InHfo ,
* 1T 2 = V2 Re^i , * 1T 3 = -V2 Im*i ,
*2s = 2* 2T 22 = 2* 2T 33 = *2 2323 = -2 Re*2 ,
*2 23 = -2Im* 2 , (A6)
* 3T 2 = V2 Re * 3 , * 3T 3 = V2 Im * 3 ,
* 4 22 = -* 4 33 = Re* 4 , vI/ 4 23 = \f432 = Im* 4 .
I

C(i)(o)(o)(i)

C(l)(0)(0)(2)
C(l)(0)(0)(3)

a

(2)(0)(0)(2)

-2Re# 2 ,

+ Re*i - Rc* 3 ,

-Im*i - Im* 3 ,

Re * 2 - ~ Rc *o - ^ Rc * 4 , (A7)

C,

(3)(0)(0)(3)

= Re* 2

~Re*c

^Re V 4
2

C(2)(0)(0)(3) = \ Im *o - - Im * 4

Explicit equations of geodesic deviation ffTTT) in _D = 4
thus take the form

A

Z (1) - 2 Re * 2 Z (1) + ( Re *i - Re * 3 ) Z (2) - ( Im *i + Im * 3 ) Z (3)

-An

t (1)(1) z« + T (1)(2) + T(1)(3) - (t (0)(0) + 1 r) z* 1

Z (2) = - Z (2) + Re * 2 Z {2) + ( Re *i - Re * 3 ) Z (1) - ~ ( Re * + Re * 4 ) Z {2) + \ (Im* - Im *4 ) Z (3)

-47T

t (2)(1) z« + r (2)(2) z( 2 ) + t (2)(3) z( 3 ) - (t (0)(0) + 1 r) z^

(A8)

Z( 3 ) = - Z (3) + Rc * 2 Z (3) - ( Im + Im * 3 ) Z (1) + i (Im* - Im * 4 ) Z (2) + - ( Rc * + Rc * 4 ) Z (3)

+47T

r (3)(1) z« + r (3)(2) z< 2 > + t (3)(3) z( 3 ) - (r (0)(0) + 1 r) z< 3 >

This fully agre es w ith the results presented in our
previous work [l29j | (after permuting the indices as
1 — > 2 3 — > 1, and changing the signs of all imaginary
parts due to a convention different from (|A1I0 .

Appendix B: Relation to other notations used in

D > 4

In the literature on higher dimensional spacetimes it
is common to use alternative conventions for the null
frame and the corresponding components. In particular,

14

in the fundamental pap ers on algebraic classification of
the Weyl tensor [72l.l73j the null frame {i,n, m,}, where
i = 2,...,D-l,

£ = m , n = mi , m 2 ,...,m D ^ 1 , (Bl)

the metric is

is employed such that
'< • i' a b>

9ab

2£( a «6) + 5*i T/'.'.'"'.' • I.e.

•n= 1
•<? = = n-n

m, = = n • nij

(B2)

Following [72|, |73| , the Weyl tensor can be decomposed
into the frame components

Cabcd = 4 Comj n{ a m\n c m : ' d ,

+ 8 Coioi n {a £ b n c m d} + 4 C oijk n {a mlm 3 c m d}
+ 4 Coioi n{JbnJd} + 8 C nj n{ a m % b i c rn 3 d y

+4C Q iij n {a £ b m l c m J d} + C ijk i m\ a rn J b m k c rn l d y
+ 8 Cwu £{ a n b £ c m d} + 4 C U jk f{ a m»^
+ AC lllj e {a ml£ c m 1 d} , (B3)

where T {abcd} = \ (T [ab][cd] + T [cd][ab] ) is a useful notation
representing the standard symmetries of the curvature
tensor. The terms in the separate lines of (|B3J) are sorted
according to their boost weight corresponding to the scal-
ing

XI.

n = A 1 n .

m, = m;

(B4)

Using (|B2I) we immediately infer that the various scalar
components in (|B3[) are explicitly given as

CoiOj

Cabcd

t a m\ l c mj ,

' =

Cabcd

£ a m\ m] mi ,

CoiOi

= Cabcd

£ a

n b £ c mi ,

Cijki =

Cabcd

abed

m i rrij m k m t

,Coioi

= Cabcd

t

n b £ d n c ,

Couj =

Cabcd

na b c d

£ n m i rrij ,

Couj

= Cabcd

t

bed

m. L n rrij ,

Clijk —

Cabcd

abed

n m i rrij m k

Cwu

= Cabcd

a rib c d

n £ n m i ,

Cuij —

Cabcd

abed

n m i n rrij .

(B5)

They are subject to a number of mutual relations which
follow from the symmetries and from the trace-free prop-
erty of the Weyl tensor, see 721 ]:

Com 1 — , Coioj — Coij 1 , Co[ij k ]
Quoi = Con 1 , Ciyki] = ,

1 /~Y k r 1 s~*l f~1 _

Mlj — —2 ik 3 ' 2 0!y ' ^Ollj -

C\[ijk] = , C\i\ = .

0.

-C

lij ,

(B6)

Now, by comparing (|B2[) with our definition (|13|) it fol-
lows that the two null frames (|B1[) and (fT"2")) are related
simply as

k

I

(B7)

Putting this identification into (|B5|) . and comparing with
(fl"4"l). we observe that

*0« =

=

Coijk ,

= ^ CoiOi

w^jfcj =

Cijki j

= - Couj

^2S =

— Coioi ,

= —Couj

tf 3 y* =

~C\ijk ,

= Cioii ,

<&4« =

(B8)

Moreover, the relations (|B6|) are equivalent to the con-
straints (|T51), (fH)l) .

Also, inglM [ll, HI the notation

^ijfe = \C\kij

= ClOli ,

(B9)

was introduced and employed, which is useful for studies
of type N and type III spacetimes, and

Co

$ = c 010 i = -±cv j

2"Coiij )

S _

1 /-« *:

2 ^ikj j

(BIO)

(where <I>^,$^,$ denote antisymmetric, symmetric
parts of $ij and its trace, respectiv ely) which is con-
venient for type D spacetimes [7^, H2 |99| . In view of
(|B8[) we thus easily identify

\t 2 «M = C^ki ,

V I' ; = 2 M-, , .

*3T-

(Bll)

Very recently, in the generalisation of the Geroch-
Held-Penrose formalism to higher dimensions [79j . an-
other convention was suggested, namely

"ij = CoiQj

{ -K, = Cuij
$ijki = C^ki

I

ijk

C

Oijk
lijk

^i — CoiOi )

% = C wu ,
(B12)

These scalars are straightforwardly related to the quan-
tities used in the present paper:

* Q1J =

W \ijk =

ty^ijkl =

= -2 $ A -

*2S =

v I / 2T i J

= ~®ij ,

y 3 i3k =

_ ijk

=

Vj-

(B13)

Appendix C: Lorentz transformations of the null
frame and the changes of

It is well known (see e.g. [z3, [iD, |93|) that gen-
eral transformations between different null frames can

15

be composed from the following simple Lorentz transfor-
mations:

• null rotation with k fixed (parametrized by D — 2
real parameters L l ):

k = k, I = l + V2L l m l + \L\ 2 k , m, = m, + %/2 Li k ,

(CI)

• null rotation with I fixed (parametrized by D — 2 real
parameters K l ):

k = k+V2K l m,, + \K\ 2 l , 1 = 1, m^m.+V^KJ ,

(C2)

• boost in the k — I plane (parametrized by a real num-
I

ber B):

k = Bk, i = B~ 1 l, rhi = m,i, (C3)

• spatial rotation in the space of (parametrized by
an orthogonal matrix ^i 3 ):

k = k , Z = I , rhi = $i j rrij , with $^ $ fc z 5ji = 5 ik .

(C4)

Due to (|13p L % — Li, K l = Ki and we employ a short-
hand \L\ 2 = L l L u \K\ 2 = K l Ki. Under these Lorentz
transformations of the frame, the Weyl scalars change
as:

• null rotation with k fixed

- 2V2~i& 0ilj L k] ,

§a«w = *2«w -2%/2(L [; * lfc]l3 -L^jjw) +4(* 04[ *L q L J - + * (y[ iL fc] L i ) ,

*25= *2S - 2V2* 1T! L ,: - 2* «L i L j ,

# 2 y = ^ 2 y + -y/2 ^ lki] L k - 2-\/2 * 1X [iL,-i - 4* k[i£j] L fc ,
# 2T « = \& 2T « + V2 ^^kjL k - \f2^ 1Ti Lj - 2^ Ql kL k L J + * ^l^| 2 >
# 3 y fc = $ 3 y fc + V2 2HjkL l — ~^23 k Li + 2L[j5 , 2T fc ] i )

+4* 1T yL fc] £i - 2 (* x *ii£ fc + * iy h£i - L ( + * 11Jfc |L| 2

+4V2^ , [j L k] L i L l - 2^ Q%{3 L k] \L\ 2 , (C5)
vj> 3Ti = vl/ 3TI + v^^yi^ - V2 {^> 2T kiL k + ^2sLi)

+2 (2V 1T jLi - * lfcJI L fc ) V - t> 1T >\L\ 2 + 2V2~y 0jh L j L k Li - \/2 4- , 3 Z7|L| 2 ,
* 4IJ - = * 4!j + 2V2 (^ 3T( iL j} - * 3(li)fc i fc )

+2y 2 iki'L k L l - 4* 2Tfc(l L i) L fc + 2*2T(«) \ L \ 2 ~ ^2sL i L J - 4* 2fc ( lJ L j) L' £

-2V2 (2^ lk i ( ,L j) L k L l + V 1(tj)k L k \L\ 2 + y lT(t L j} \L\ 2 - 2V 1Tk L k L l L j )

+A^> Qk iL k L l L l L ] - A^ Qk{ ,L 3) L k \L\ 2 + |i| 4 ,

• null rotation with I fixed

+2* 21fc ., ; X' £ if / - 4if (l * 2TJ)fc i<:' £ + 2* 2T(y) I if | 2 - 2^ 2S K i K j + W> 2Ki K j) K k
-2V2(2^ 3k i ( ,K 3) K k K l + V 3i i j)k K k \K\ 2 + * 3T(l i^) |if| 2 - 2V 3Tk K k K i K j )
+W 4k ,K k K l K, l K j - W 4H ,K f) K k \K\ 2 + ^^\K\ 4 ,

+4^^^ - 2 (*3, M /ir fc + * 3 y^i - * 3 "^j)^ + *3«*l#l 2

W^^iQi^ - 2V2~y ilU K k] \K\ 2 , (C6)
+2 (2V 3T iKi - y 3kJ *K k ) K j - ^T'l^l 2 + 2V2 ^ k K j K k K { - V^^^K^K] 2 ,

16

*2S = *2S - 2V2* 3TI ^ 4 - 2* 4 , ti K i K> ,
>i' ; = >!' ; ,

• boost in the k — I plane

= i? , * 1T > = 5 * 1T . ,

# 2<JW = * 2<JW , § 2 y = tf 2 „ , * 2S = * 2S , * 2T13 = * 2r y , (C7)

§ 4IJ = B- 2 * 4I3 ,

• spatial rotation in the space of

= $/ *i TP ,

^ 2I3fc!

- ti>. n

* °

:/

$ fc p $ ; «* 2 „op 9 ,

= *25 ,

*/

— $.°

z

$ P

*/

W4P9 .

r

Emparan R and Reall H S 2008 Black holes [10
in higher dimensions Living. Rev. Relativity 11 6
http: / / www . livingreviews .org / lrr- 200 8^6]

Emparan R and Reall H S 2006 Black rings Class. Quan- [11
turn Grav. 23 R169-R197

Obers N A 2009 Black holes in higher-dimensional grav-
ity Lect. Notes Phys. 769 211-258 [12;
Kanti P 2004 Black holes in theories with large extra
dimensions: a review Int. J. Mod. Phys. A 19 4899-
4951

Gibbons G W, Lii H, Page D N and Pope C N 2004 [13;

Rotating black holes in higher dimensions with a cos-

mological constant Phys. Rev. Lett. 93 171102 (4pp)

Chen W, Lii H and Pope C N 2006 General Kerr-NUT- [14

AdS metrics in all dimensions Class. Quantum Grav. 23

5323-5340

Krtous P, Frolov V P and Kubiznak D 2008 Hidden sym- [15
metries of higher-dimensional black holes and unique-
ness of the Kerr-NUT-(A)dS spacetime Phys. Rev. D
78 064022 (5pp)
Berti E, Cardoso V and Starinets A O 2009 Quasinormal [16
modes of black holes and black branes Class. Quantum
Grav. 26 163001 (108pp) ' [17

Emparan R and Reall H S 2002 Generalized Weyl solu-
tions Phys. Rev. D 65 084025 (26pp) [18;

Harmark T 2004 Stationary and axisymmetric solutions
of higher-dimensional general relativity Phys. Rev. D 70
124002 (25pp)

Charmousis C and Gregory R 2004 Axisymmetric met-
rics in arbitrary dimensions Class. Quantum Grav. 21
527-553

Mishima T and Iguchi H 2006 New axisymmetric sta-
tionary solutions of five-dimensional vacuum Einstein
equations with asymptotic flatness Phys. Rev. D 73
044030 (6pp)

Giusto S and Saxena A 2007 Stationary axisymmet-
ric solutions of five-dimensional gravity Class. Quantum
Grav. 24 4269-4294

Coimbra-Araujo C H and Letelier P S Thin disk in
higher dimensional space-time and dark matter inter-
pretation Phys. Rev. D 76 043522 (17pp)
Godazgar M and Reall H S 2009 Algebraically spe-
cial axisymmetric solutions of the higher-dimensional
vacuum Einstein equation Class. Quantum Grav. 26
165009 (29pp)

Myers R C 1987 Higher-dimensional black holes in com-
pactified space-times Phys. Rev. D 35 455-466
London L A J 1995 Arbitrary dimensional cosmological
multi-black holes Nucl. Phys. B 434 709-735
Gibbons G W, Horowitz G T and Townsend P K 1995

17

Higher-dimensional resolution of dilatonic black-hole
singularities Class. Quantum Grav. 12 297-317

[19] Welch D L 1995 Smoothness of the horizons of multi-
black-hole solutions Phys. Rev. D 52 985-991

[20] Ivashchuk V D and Melnikov V N 2001 Exact solutions
in multidimensional gravity with antisymmetric forms
Class. Quantum Grav. 18 R87-R152

[21] Astefanesei D, Mann R B and Radu E 2004 Reissner-
Nordstrom-de Sitter black hole, planar coordinates and
dS/CFT J. High Energy Phys. JHEP01(2004)029

[22] Candlish G N and Reall H S 2007 On the smooth-
ness of static multi-black hole solutions of higher dimen-
sional Einstein-Maxwell theory Class. Quantum Grav.
24 6025-6039

[23] Ida D, Ishihara H, Kimura M, Matsuno K, Morisawa Y
and Tomizawa S 2007 Cosmological black holes on
Taub-NUT space in five-dimensional Einstein-Maxwell
theory Class. Quantum Grav. 24 3141-3149

[24] Horowitz G T and Myers R C 1999 AdS-CFT correspon-
dence and a new positive energy conjecture for general
relativity Phys. Rev. D 59 026005 (12pp)

[25] Galloway G J, Surya S and Woolgar E 2002 A unique-
ness theorem for the anti-de Sitter soliton Phys. Rev.
Lett. 88 101102 (4pp)

[26] Sanoglu 6 and Tekin B 2009 Comment on 'A sec-
ond anti-de Sitter universe' Class. Quantum Grav. 26
048001 (2pp)

[27] Sanoglu and Tekin B 2009 Note on cosmological Levi-
Civita spacetimes in higher dimensions Phys. Rev. D 79
087502 (2pp)

[28] Griffiths J B and Podolsky J 2010 The Linet-Tian so-
lution with a positive cosmological constant in four and
higher dimensions Phys. Rev. D 81 064015 (6pp)

[29] Wiseman T 2003 Static axisymmetric vacuum solutions
and non-uniform black strings Class. Quantum Grav.
20 1137-1175

[30] Kleihaus B, Kunz J and Radu E 2006 New nonuni-
form black string solutions J. High Energy Phys.
JHEP06(2006)016

[31] Copsey K and Horowitz G T 2006 Gravity dual of
gauge theory on S 2 x S 1 x R J. High Energy Phys.
JHEP06(2006)021

[32] Mann R B, Radu E and Stelea C 2006 Black string
solutions with negative cosmological constant J. High
Energy Phys. JHEP09(2006)073

[33] Zhao L, Niu K, Xia B, Dou Y and Ren J 2007 Non-
uniform black strings with Schwarzschild-(anti-)de Sit-
ter foliation Class. Quantum Grav. 24 4587-4599

[34] Brihaye Y and Delsate T 2007 Charged-rotating black
holes and black strings in higher dimensional Einstein-
Maxwell theory with a positive cosmological constant
Class. Quantum Grav. 24 4691-4709

[35] Brihaye Y, Radu E and Stelea C 2007 Black strings with
a negative cosmological constant: inclusion of electric
charge and rotation Class. Quantum Grav. 24 4839-
4869

[36] Brihaye Y, Kunz J and Radu E 2009 From black strings

to black holes: nuttier and squashed AdSs solutions

J. High Energy Phys. JHEP08(2009)025
[37] Gregory R and Laflamme R 1993 Black strings and p-

branes are unstable Phys. Rev. Lett. 70 2837-2840
[38] Horowitz G T and Maeda K 2001 Fate of the black string

instability Phys. Rev. Lett. 87 131301 (4pp)

[39] Harmark T, Niarchos V and Obers N A 2007 Instabil-
ities of black strings and branes Class. Quantum Grav.
24 R1-R90

[40] Podolsky J and Ortaggio M 2006 Robinson-Trautman
spacetimes in higher dimensions Class. Quantum Grav.
23 5785-5797

[41] Ortaggio M, Podolsky J and Zofka M 2008 Robinson-
Trautman spacetimes with an electromagnetic field in
higher dimensions Class. Quantum Grav. 25 025006
(18pp)

[42] Ortaggio M, Pravda V and Pravdova A 2009 Higher

dimensional Kerr-Schild spacetimes Class. Quantum

Grav. 26 025008 (28pp)
[43] Malek T and Pravda V 2011 Kerr-Schild spacetimes

with an (A)dS background Class. Quantum Grav. 28

125011 (17pp)

[44] Giirses M and Sanoglu 2002 Accelerated charged
Kerr-Schild metrics in D dimensions Class. Quantum
Grav. 19 4249-4261

[45] Podolsky J 2011 Photon rockets moving arbitrarily in
any dimension Int. J. Mod. Phys. D 20 335-360

[46] Cardoso V, Dias 6 J C and Lemos J P S 2004 Nariai,
Bertotti-Robinson, and anti-Nariai solutions in higher
dimensions Phys. Rev. D 70 024002 (lOpp)

[47] Binetruy P, Deffayet C, Ellwanger U and Langlois D
2000 Brane cosmological evolution in a bulk with cos-
mological constant Phys. Lett. B 477 285-291

[48] Shiromizu T, Maeda K and Sasaki M 2000 The Einstein
equations on the 3-brane world Phys. Rev. D 62 024012

[49] Brax P, van den Bruck C and Davis A-C 2004 Brane
world cosmology Rep. Prog. Phys. 67 2183-2231

[50] Garcia A A and Carlip S 2007 n-dimensional generaliza-
tions of the Friedmann-Robertson-Walker cosmology
Phys. Lett. B 645 101-107

[51] Maartens R and Koyama K 2010 Brane-
world gravity Living. Rev. Relativity 13 5
http: / /www.livingreviews.org/lrr-2010-5

[52] Hervik S 2002 Multidimensional cosmology: spatially
homogeneous models of dimension 4+1 Class. Quantum
Grav. 19 5409-5427

[53] Clarkson R and Mann R B 2006 Soliton solutions to the
Einstein equations in five dimensions Phys. Rev. Lett.
96 051104 (4pp)

[54] Clarkson R and Mann R B 2006 Eguchi-Hanson solitons
in odd dimensions Class. Quantum Grav. 23 1507-1523

[55] Podolsky J and Zofka M 2009 General Kundt spacetimes
in higher dimensions Class. Quantum Grav. 26 105008
(18pp)

[56] Coley A, Hervik S, Papadopoulos G and Pelavas N 2009
Kundt spacetimes Class. Quantum Grav. 26 105016
(34pp)

[57] Brinkmann H W 1925 Einstein spaces which are mapped
conformally on each other Math. Annal. 94 119-145

[58] Gibbons G W and Ruback P J 1986 Classical gravitons
and their stability in higher dimensions Phys. Lett. B
171 390-395

[59] Coley A, Milson R, Pelavas N, Pravda V, Pravdova A
and Zalaletdinov 2003 Generalizations of pp-wave space-
times in higher dimensions Phys. Rev. D 67 104020
(4pp)

[60] Coley A, Hervik S 2004 Brane waves Class. Quantum

Grav. 21 5759-5766
[61] Ortm T 2004 Gravity and Strings (Cambridge: Cam-

18

bridge University Press)

Coley A, Milson R, Pravda V and Pravdova A 2004 Van- [84
ishing scalar invariant spacetimes in higher dimensions
Class. Quantum Grav. 21 5519-5542

Coley A, Fuster A, Hervik S and Pelavas N 2006 Higher [85;
dimensional VSI spacetimes Class. Quantum Grav. 23
7431-7444 " ' [86;

Giirses M and Karasu A 2001 Some higher-dimensional
vacuum solutions Class. Quantum Grav. 18 509-516
Coley A, Hervik S and Pelavas N 2006 On spacetimes [87
with constant scalar invariants Class. Quantum Grav.
23 3053-3074

Frolov V P and Fursaev D V 2005 Gravitational field [J
of a spinning radiation beam pulse in higher dimensions
Phys. Rev. D 71 104034 (16pp)

Frolov V P, Israel W and Zelnikov A 2005 Gravitational [89;
field of relativistic gyratons Phys. Rev. D 72 084031
(llpp)

Frolov V P and Zelnikov A 2005 Relativistic gyratons in [90
asymptotically AdS spacetime Phys. Rev. D 72 104005
(10pp)

Frolov V P and Zelnikov A 2006 Gravitational field of [91
charged gyratons Class. Quantum Grav. 23 2119-2128
Frolov V P and Lin F 2006 String gyratons in super-
gravity Phys. Rev. D 73 104028 (7pp) [92;
Caldarelli M M, Klemm D and Zorzan E 2007 Super-
symmetric gyratons in five dimensions Class. Quantum
Grav. 24 1341-1357 [93
Coley A 2008 Classification of the Weyl tensor in higher
dimensions and applications Class. Quantum Grav. 25
033001 (29pp) [94
Coley A, Milson R, Pravda V and Pravdova A 2004
Classification of the Weyl tensor in higher dimensions
Class. Quantum Grav. 21 L35-L42 [95
Ortaggio M 2009 Bel-Debever criteria for the classifi-
cation of the Weyl tensor in higher dimensions Class.
Quantum Grav. 26 195015 (8pp) [96
Senovilla J M M 2010 Algebraic classification of the
Weyl tensor in higher dimensions based on its 'superen-
ergy' tensor Class. Quantum Grav. 27 222001 (7pp) [97
Pravda V, Pravdova A, Coley A, and Milson R 2004
Bianchi identities in higher dimensions Class. Quantum
Grav. 21 2873-2898 [98
Ortaggio M, Pravda V and Pravdova A 2007 Ricci iden-
tities in higher dimensions Class. Quantum Grav. 24
1657-1664 [99
Pravda V and Pravdova A 2008 The Newman-Penrose
formalism in higher dimensions: vacuum spacetimes
with a non-twisting geodetic multiple Weyl aligned null [100
direction Class. Quantum Grav. 25 235008 (27pp)
Durkee M, Pravda V, Pravdova A and Reall H S 2010 [101
Generalization of the Geroch-Held-Penrose formalism
to higher dimensions Class. Quantum Grav. 27 215010
(21pp) ' [102;

Durkee M and Reall H S 2011 Perturbations of
higher-dimensional spacetimes Class. Quantum Grav.
28 035011 (20pp)

Pravda V, Pravdova A and Ortaggio M 2007 Type D
Einstein spacetimes in higher dimensions Class. Quan- [103
turn Grav. 24 4407-4428

Durkee M 2009 Type II Einstein spacetimes in higher [104
dimensions Class. Quantum Grav. 26 195010 (lOpp)
Ortaggio M, Pravda V and Pravdova A 2010 Type III [105
and N Einstein spacetimes in higher dimensions: Gen-

eral properties Phys. Rev. D 82 064043 (16pp)
Ortaggio M, Pravda V and Pravdova A 2011 On higher
dimensional Einstein spacetimes with a warped extra
dimension Class. Quantum Grav. 28 105006 (17pp)
Marolf D and Ross S F 2002 Plane waves: to infinity
and beyond! Class. Quantum Grav. 19 6289-62302
Cardoso V, Dias 6 J C and Lemos J P S 2003 Grav-
itational radiation in D-dimensional spacetimes Phys.
Rev. D 67 064026 (13pp)

Hollands S and Wald R M 2004 Conformal null infinity
does not exist for radiating solutions in odd spacetime
dimensions Class. Quantum Grav. 21 5139-5145
Hollands S and Ishibashi A 2005 Asymptotic flatness
and Bondi energy in higher dimensional gravity J. Math.
Phys. 46 022503 (31pp)

Ishibashi A 2008 Higher dimensional Bondi energy with
a globally specified background structure Class. Quan-
tum Grav. 25 165004 (19pp)

Pravdova A, Pravda V and Coley A 2005 A note on the
peeling theorem in higher dimensions Class. Quantum
Grav. 22 2535-2538

Alesci E and Montani G 2005 Can gravitational waves
be markers of an extra-dimension? Int. J. Mod. Phys. D
14 923-931

Krtous P and Podolsky J 2004 Asymptotic directional
structure of radiative fields in spacetimes with a cosmo-
logical constant Class. Quantum Grav. 21 R233-R273
Krtous P and Podolsky J 2006 Asymptotic structure of
radiation in higher dimensions Class. Quantum Grav.
23 1603-1615

Anderson M T and Chrusciel P T 2005 Asymptotically
simple solutions of the vacuum Einstein equations in
even dimensions Commun. Math. Phys. 260 557-577
Bizori P, Chmaj T and Schmidt B G 2005 Critical be-
havior in vacuum gravitational collapse in 4+1 dimen-
sions Phys. Rev. Lett. 95 071102 (4pp)
Choquet-Bruhat Y, Chrusciel P T and Loizelet J 2006
Global solutions of the Einstein-Maxwell equations in
higher dimensions Class. Quantum Grav. 23 7383-7394
Beig R and Chrusciel P T 2007 The asymptotics of sta-
tionary electro- vacuum metrics in odd spacetime dimen-
sions Class. Quantum Grav. 24 867-874
Mironov A D and Morozov A Y 2008 Radiation be-
yond four space-time dimensions Theor. Math. Phys.
156 1209-1217

Ortaggio M, Pravda V and Pravdova A 2009 Asymp-
totically flat, algebraically special spacetimes in higher
dimensions Phys. Rev. D 80 084041 (5pp)
Levi-Civita T 1926 Sur l'ecart geodesique Math. Ann.
97 291-320

Synge J L 1926 The first and second variations of the
length integral in Riemannian space Proc. Lond. Math.
Soc. 25 247-264

Synge J L 1934 On the deviation of geodesies and null-
geodesics, particularly in relation to the properties of
spaces of constant curvature and indefinite line-element
Ann. Math. 35 705-713; reprinted in Gen. Rel. Grav.
41 (2009) 1205-1214

Synge J L and Schild A 1949 Tensor Calculus (Toronto:
University of Toronto Press)

Trautman A 2009 Editorial note Gen. Rel. Grav. 41
1195-1203

Pirani F A E 1956 On the physical significance of
the Riemann tensor Acta Phys. Polon. 15 389-405;

19

reprinted in Gen. Rel. Grav. 41 (2009) 1215-1232

[106] Pirani F A E 1957 Invariant formulation of gravitational
radiation theory Phys. Rev. 105 1089-1099

[107] Bondi H, Pirani F A E and Robinson I 1959 Gravita-
tional waves in general relativity III. Exact plane waves
Proc. Roy. Soc. A 251 519-533

[108] Sachs R K 1960 Propagation laws for null and type III
gravitational waves Z. Phystk 157 462-477

[109] Weber J 1960 Detection and generation of gravitational
waves Phys. Rev. 117 306-313

[110] Synge J L 1960 Relativity: The General Theory (Ams-
terdam: North-Holland)

[111] Weber J 1961 General Relativity and Gravitational
Waves (New York: Interscience)

[112] Ehlers J and Kundt W 1962 Exact solutions of the gravi-
tational field equations, in Gravitation: An Introduction
to Current Research, L. Witten (ed.) (New York: Wiley)
pp 49-101

[113] Pirani F A E 1965 Introduction to gravitational radi-
ation theory, in Brandeis Lectures on General Relativ-
ity, S. Deser and K. W. Ford (eds.) (Englewood Cliffs:
Prentice-Hall) pp 249-373

[114] Szekeres P 1965 The gravitational compass J. Math.
Phys. 6 1387-1391

[115] Misner C W, Thorne K S and Wheeler J A 1973 Grav-
itation (San Francisco: Freeman)

[116] Wald R M 1984 General Relativity (Chicago: University
of Chicago Press)

[117] Hobson M P, Efstathiou G P and Lasenby A N
2006 General Relativity: An Introduction for Physicists
(Cambridge: Cambridge University Press)

[118] de Felice F and Bini D 2010 Classical Measurements in
Curved Space-Times (Cambridge: Cambridge Univer-
sity Press)

[119] Hodgkinson D E 1972 A modified equation of geodesic
deviation Gen. Rel. Grav. 3 351-375

[120] Mashhoon B 1975 On tidal phenomena in a strong grav-
itational field Astrophys. J. 197 705-716

[121] Mashhoon B 1977 Tidal radiation Astrophys. J. 216
591-609

[122] Bazanski S L 1977 Kinematics of relative motion of test
particles in general relativity Ann. Inst. Henri Poincare
27 115-144

[123] Aleksandrov A N and Piragas K A 1979 Geodesic struc-
ture I. Relative dynamics of geodesies Theor. Math.
Phys. 38 48-56

[124] Li W Q and Ni W T 1979 Coupled inertial and gravita-
tional effects in the proper reference frame of an accel-
erated, rotating observer J. Math. Phys. 20 1473-1480

[125] Audretsch J and Lammerzahl C 1983 Local and non-
local measurements of the Riemann tensor Gen. Rel.
Grav. 15 495-510

[126] Ciufolini I 1986 Generalized geodesic deviation equation
Phys. Rev. D 34 1014-1017

[127] Ciufolini I and Demianski M 1986 How to measure the
curvature of space-time Phys. Rev. D 34 1018-1020

[128] Chicone C and Mashhoon B 2002 The generalized Ja-
cobi equation Class. Quantum Grav. 19 4231-4248

[129] Bicak J and Podolsky J 1999 Gravitational waves in
vacuum spacetimes with cosmological constant. II. De-
viation of geodesies and interpretation of nontwisting
type N solutions J. Math. Phys. 40 4506-4517

[130] Balakin A, van Holten J W and Kerner R 2000 Mo-
tions and worldline deviations in Einstein-Maxwell the-
ory Class. Quantum Grav. 17 5009-5023

[131] Manoff S 2001 Deviation operator and deviation equa-
tions over spaces with affine connections and metrics
J. Geom. Phys. 39 337-350

[132] Kerner R, van Holten J W and Colistete R Jr 2001 Rel-
ativistic epicycles: another approach to geodesic devia-
tions Class. Quantum Grav. 18 4725-4742

[133] Colistete R Jr, Leygnac C and Kerner R 2002 Higher-
order geodesic deviations applied to the Kerr metric
Class. Quantum Grav. 19 4573-4589

[134] Chicone C and Mashhoon B 2006 Tidal dynamics in
Kerr spacetime Class. Quantum Grav. 23 4021-4033

[135] Mullari T and Tammelo R 2006 On the relativistic tidal
effects in the second approximation Class. Quantum
Grav. 23 4047-4067

[136] Jordan P, Ehlers J and Kundt W 1960 Strenge Losungen
der Feldgleichungen der Allgemeinen Relativitatstheorie
Akad. Wiss. Lit. Mainz, Abhandl. Math.-Nat. Kl. nr. 2
(85pp)

[137] Sachs R 1961 Gravitational waves in general relativity.
VI. The outgoing radiation condition Proc. Roy. Soc. A
264 309-338

[138] Newman E and Penrose R 1962 An approach to gravita-
tional radiation by a method of spin coefficients J. Math.
Phys. 3 566-578; 4 (1963) 998

[139] Penrose R and Rindler W 1984, 1986 Spinors and Space-
Time (Cambridge: Cambridge University Press)

[140] Griffiths J B 1991 Colliding Plane Waves in General
Relativity (Oxford: Oxford University Press)

[141] Stephani H, Kramer D, MacCallum M, Hoenselaers C
and Herlt E 2003 Exact Solutions of Einstein's Field
Equations (Cambridge: Cambridge University Press)

[142] Griffiths J B and Podolsky J 2009 Exact Space-Times in
Einstein's General Relativity (Cambridge: Cambridge
University Press)